Jim,
In addition to the five series 1 meg resistors from HV+, there's a 6th
resistor, a 15K 1/2W in parallel with the panel multimeter. However, the 15K
resistor is across the multimeter only when the meter mode switch is in the HV
position -- otherwise it would affect other readings. The combination of the
meter in parallel with the 15K resistor is in series with all 5 string
resistors on the cold end. The meter's low internal R swamps the 15K resistor.
In fact, I confirmed it by briefly removing the 15K resistor from the circuit
while testing in the HV position and the meter read little change. It would
appear the primary reason for the 15K resistor is to float-down the HV on the
multimeter switch for safety and eliminate any HV contact arcing when switching
between meter positions. Here's how I reached the 4.5 meg value:
5KV is full scale on the 1 mA movement. But I need the equivalent of 1.1 mA
for full scale (FS) because the meter is reading 500V low. That's a meter
current discrepancy of 0.1 mA as each 1KV change is 0.2 mA. The meter legend
is nearly linear between KV markings as expected. Neglecting the meter's small
internal R, 5KV/1.1 mA = 4.54 meg instead of 5.0 meg as designed. I ordered
several values so that I end up as close as possible. I don’t care if the
meter is really 1.0 mA FS or not. I just know that an additional 0.1 mA is
needed. If the meter is linear, then the computation should be reasonably
close.
I may as well document the internal meter resistance, using a series R and a
pot as a rheostat in parallel across the meter. With a bench DC power supply,
adjust series R for full scale, then exactly half scale with the rheostat.
Remove rheostat and measure it. The result is the meter's internal DC
resistance.
Paul, W9AC
-----Original Message-----
From: MU 4CX250B [mailto:4cx250b@miamioh.edu]
Sent: Thursday, January 12, 2017 4:24 PM
To: Paul Christensen <w9ac@arrl.net>
Cc: Amps group <amps@contesting.com>
Subject: Re: [Amps] Alpha Seventy HV Meter Readings
To compute the value of the bottom resistor in the divider string, don't you
need to compute the parallel resistance of the bottom resistor and the
resistance in series with the 1mA panel meter?
Presumably, the internal resistance of the meter is low enough to ignore.
73,
Jim w8zr
Sent from my iPhone
> On Jan 12, 2017, at 2:09 PM, Paul Christensen <w9ac@arrl.net> wrote:
>
> I've recently acquired an air-cooled version of the Alpha Seventy amplifier.
> I notice that HV reads about 500V low when comparing it to a Fluke DMM
> with Fluke HV probe. The supply is the same as my vapor-cooled
> version. The vapor amp's HV meter tracks in line with the Fluke.
>
> My first thought was that the meter was out of calibration. It's a
> common 1 mA DC movement. A NOS replacement meter also reads low by
> 500V. That leaves the five 1-meg HV metering divider resistors but
> they all measure within 1% tolerance as they should -- metal film
> resistors rarely drift out of spec.
>
> So, my plan is to change the value of only the 1-meg resistor on the
> cold end of the 5 megohm string. By computation, I get 500K instead
> of 1-meg for a total string value of 4.5-meg instead of 5.0-meg. Seem
> like a reasonable approach to y'all?
>
> Paul, W9AC
>
>
>
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|